update pre-commit

.pre-commit-config.yaml

@@ -41,4 +41,4 @@
         description: Check C++ code style using cpplint.py.
         entry: bash ./.travis/codestyle/cpplint_pre_commit.hook
         language: system
-        files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx)$
+        files: \.(c|cc|cxx|cpp|cu|h|hpp|hxx)$

.travis/codestyle/cpplint_pre_commit.hook

@@ -24,4 +24,4 @@ for file in $files; do
     fi
 done
 
-exit $TOTAL_ERRORS
+exit $TOTAL_ERRORS

.travis/codestyle/pylint_pre_commit.hook

@@ -0,0 +1,25 @@
+#!/bin/bash
+
+TOTAL_ERRORS=0
+
+
+DIR="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
+export PYTHONPATH=$DIR:$PYTHONPATH
+
+readonly VERSION="2.12.0"
+version=$(pylint --version | grep 'pylint')
+
+if ! [[ $version == *"$VERSION"* ]]; then
+    pip install pylint==2.12.0
+fi
+
+# The trick to remove deleted files: https://stackoverflow.com/a/2413151
+for file in $(git diff --name-status | awk '$1 != "D" {print $2}'); do
+    pylint --disable=all --load-plugins=docstring_checker \
+    --enable=doc-string-one-line,doc-string-end-with,doc-string-with-all-args,doc-string-triple-quotes,doc-string-missing,doc-string-indent-error,doc-string-with-returns,doc-string-with-raises $file;
+    TOTAL_ERRORS=$(expr $TOTAL_ERRORS + $?);
+done
+
+exit $TOTAL_ERRORS
+#For now, just warning:
+#exit 0

.travis/precommit.sh

@@ -1,4 +1,19 @@
 #!/bin/bash
+
+# Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+# 
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+# 
+#     http://www.apache.org/licenses/LICENSE-2.0
+# 
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 function abort(){
     echo "Your commit not fit PaddlePaddle code style" 1>&2
     echo "Please use pre-commit scripts to auto-format your code" 1>&2

applications/Automatic_label/automatic_label.py

@@ -15,28 +15,24 @@
 import os
 from dataclasses import dataclass, field
 
-import paddle
+import matplotlib.pyplot as plt
+import nltk
 import numpy as np
-import requests
-from paddlenlp.trainer import PdArgumentParser
-from PIL import Image
-
 import paddle
 import paddle.nn.functional as F
+import requests
+from paddlenlp.trainer import PdArgumentParser
+from paddlenlp.transformers import AutoTokenizer
 from PIL import Image, ImageDraw, ImageFont
 
-from paddlemix.processors.groundingdino_processing import GroudingDinoProcessor
+from paddlemix.models.blip2.modeling import Blip2ForConditionalGeneration
 from paddlemix.models.groundingdino.modeling import GroundingDinoModel
 from paddlemix.models.sam.modeling import SamModel
+from paddlemix.processors.blip_processing import (
+    Blip2Processor, BlipImageProcessor, BlipTextProcessor)
+from paddlemix.processors.groundingdino_processing import GroudingDinoProcessor
 from paddlemix.processors.sam_processing import SamProcessor
-from paddlenlp.transformers import AutoTokenizer
-from paddlemix.processors.blip_processing import BlipImageProcessor, BlipTextProcessor
-from paddlemix.models.blip2.modeling import Blip2ForConditionalGeneration
-from paddlemix.processors.blip_processing import Blip2Processor
-import nltk
-
 from paddlemix.utils.log import logger
-import matplotlib.pyplot as plt
 
 
 def show_mask(mask, ax, random_color=False):
@@ -54,7 +50,7 @@ def show_box(box, ax, label):
     w, h = box[2] - box[0], box[3] - box[1]
     ax.add_patch(
         plt.Rectangle(
-            (x0, y0), w, h, edgecolor='green', facecolor=(0, 0, 0, 0), lw=2))
+            (x0, y0), w, h, edgecolor="green", facecolor=(0, 0, 0, 0), lw=2))
     ax.text(x0, y0, label)
 
 
@@ -71,15 +67,16 @@ class DataArguments:
 
     prompt: str = field(
         default="describe the image",
-        metadata={"help": "The prompt of the image to be generated."
-                  })  # "Question: how many cats are there? Answer:"
+        metadata={"help": "The prompt of the image to be generated."},
+    )  # "Question: how many cats are there? Answer:"
 
 
 @dataclass
 class ModelArguments:
     """
     Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
     """
+
     blip2_model_name_or_path: str = field(
         default="paddlemix/blip2-caption-opt2.7b",
         metadata={"help": "Path to pretrained model or model identifier"}, )
@@ -125,24 +122,24 @@ def generate_caption(raw_image, prompt, processor, blip2_model):
 def generate_tags(caption):
     lemma = nltk.wordnet.WordNetLemmatizer()
 
-    nltk.download(['punkt', 'averaged_perceptron_tagger', 'wordnet'])
+    nltk.download(["punkt", "averaged_perceptron_tagger", "wordnet"])
     tags_list = [
         word for (word, pos) in nltk.pos_tag(nltk.word_tokenize(caption))
-        if pos[0] == 'N'
+        if pos[0] == "N"
     ]
     tags_lemma = [lemma.lemmatize(w) for w in tags_list]
-    tags = ', '.join(map(str, tags_lemma))
+    tags = ", ".join(map(str, tags_lemma))
 
     return tags
 
 
 def main():
     parser = PdArgumentParser((ModelArguments, DataArguments))
     model_args, data_args = parser.parse_args_into_dataclasses()
-    url = (data_args.input_image)
+    url = data_args.input_image
 
     logger.info("blip2_model: {}".format(model_args.blip2_model_name_or_path))
-    #bulid blip2 processor
+    # bulid blip2 processor
     blip2_tokenizer_class = AutoTokenizer.from_pretrained(
         model_args.text_model_name_or_path, use_fast=False)
     blip2_image_processor = BlipImageProcessor.from_pretrained(
@@ -162,27 +159,27 @@ def main():
     logger.info("blip2_model build finish!")
 
     logger.info("dino_model: {}".format(model_args.dino_model_name_or_path))
-    #bulid dino processor
+    # bulid dino processor
     dino_processor = GroudingDinoProcessor.from_pretrained(
         model_args.dino_model_name_or_path)
-    #bulid dino model
+    # bulid dino model
     dino_model = GroundingDinoModel.from_pretrained(
         model_args.dino_model_name_or_path)
     dino_model.eval()
     logger.info("dino_model build finish!")
 
-    #buidl sam processor
+    # buidl sam processor
     sam_processor = SamProcessor.from_pretrained(
         model_args.sam_model_name_or_path)
-    #bulid model
+    # bulid model
     logger.info("SamModel: {}".format(model_args.sam_model_name_or_path))
     sam_model = SamModel.from_pretrained(
         model_args.sam_model_name_or_path, input_type="boxs")
     logger.info("SamModel build finish!")
 
-    #read image
+    # read image
     if os.path.isfile(url):
-        #read image
+        # read image
         image_pil = Image.open(data_args.input_image)
     else:
         image_pil = Image.open(requests.get(url, stream=True).raw)
@@ -191,26 +188,26 @@ def main():
         image_pil,
         prompt=data_args.prompt,
         processor=blip2_processor,
-        blip2_model=blip2_model)
+        blip2_model=blip2_model, )
 
     det_prompt = generate_tags(caption)
     logger.info("det prompt: {}".format(det_prompt))
 
     image_pil = image_pil.convert("RGB")
 
-    #preprocess image text_prompt
+    # preprocess image text_prompt
     image_tensor, mask, tokenized_out = dino_processor(
         images=image_pil, text=det_prompt)
 
     with paddle.no_grad():
         outputs = dino_model(
             image_tensor,
             mask,
-            input_ids=tokenized_out['input_ids'],
-            attention_mask=tokenized_out['attention_mask'],
+            input_ids=tokenized_out["input_ids"],
+            attention_mask=tokenized_out["attention_mask"],
             text_self_attention_masks=tokenized_out[
-                'text_self_attention_masks'],
-            position_ids=tokenized_out['position_ids'])
+                "text_self_attention_masks"],
+            position_ids=tokenized_out["position_ids"], )
 
     logits = F.sigmoid(outputs["pred_logits"])[0]  # (nq, 256)
     boxes = outputs["pred_boxes"][0]  # (nq, 4)
@@ -265,12 +262,12 @@ def main():
             show_box(box, plt.gca(), label)
 
         plt.title(caption)
-        plt.axis('off')
+        plt.axis("off")
         plt.savefig(
-            os.path.join(model_args.output_dir, 'mask_pred.jpg'),
+            os.path.join(model_args.output_dir, "mask_pred.jpg"),
             bbox_inches="tight",
             dpi=300,
-            pad_inches=0.0)
+            pad_inches=0.0, )
 
     logger.info("finish!")
 

applications/CVinW/grounded_sam.py

@@ -1,21 +1,35 @@
-from dataclasses import dataclass, field
+# Copyright (c) 2023 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
 import os
 import sys
-import numpy as np
-import requests
+from dataclasses import dataclass, field
 from typing import List
 
+import matplotlib.pyplot as plt
+import numpy as np
 import paddle
 import paddle.nn.functional as F
-from PIL import Image, ImageDraw, ImageFont
+import requests
 from paddlenlp.trainer import PdArgumentParser
-from paddlemix.utils.log import logger
+from PIL import Image, ImageDraw, ImageFont
 
-from paddlemix.processors.groundingdino_processing import GroudingDinoProcessor
 from paddlemix.models.groundingdino.modeling import GroundingDinoModel
 from paddlemix.models.sam.modeling import SamModel
+from paddlemix.processors.groundingdino_processing import GroudingDinoProcessor
 from paddlemix.processors.sam_processing import SamProcessor
-import matplotlib.pyplot as plt
+from paddlemix.utils.log import logger
 
 
 def show_mask(mask, ax, random_color=False):
@@ -33,7 +47,7 @@ def show_box(box, ax, label):
     w, h = box[2] - box[0], box[3] - box[1]
     ax.add_patch(
         plt.Rectangle(
-            (x0, y0), w, h, edgecolor='green', facecolor=(0, 0, 0, 0), lw=2))
+            (x0, y0), w, h, edgecolor="green", facecolor=(0, 0, 0, 0), lw=2))
     ax.text(x0, y0, label)
 
 
@@ -80,44 +94,44 @@ class ModelArguments:
 def main():
     parser = PdArgumentParser((ModelArguments, DataArguments))
     model_args, data_args = parser.parse_args_into_dataclasses()
-    url = (data_args.input_image)
-    #bulid dino processor
+    url = data_args.input_image
+    # bulid dino processor
     dino_processor = GroudingDinoProcessor.from_pretrained(
         model_args.dino_model_name_or_path)
 
-    #bulid dino model
+    # bulid dino model
     logger.info("dino_model: {}".format(model_args.dino_model_name_or_path))
     dino_model = GroundingDinoModel.from_pretrained(
         model_args.dino_model_name_or_path)
     dino_model.eval()
-    #buidl sam processor
+    # buidl sam processor
     sam_processor = SamProcessor.from_pretrained(
         model_args.sam_model_name_or_path)
-    #bulid model
+    # bulid model
     logger.info("SamModel: {}".format(model_args.sam_model_name_or_path))
     sam_model = SamModel.from_pretrained(
         model_args.sam_model_name_or_path, input_type="boxs")
 
-    #read image
+    # read image
     if os.path.isfile(url):
-        #read image
+        # read image
         image_pil = Image.open(url).convert("RGB")
     else:
         image_pil = Image.open(requests.get(url, stream=True).raw).convert(
             "RGB")
-    #preprocess image text_prompt
+    # preprocess image text_prompt
     image_tensor, mask, tokenized_out = dino_processor(
         images=image_pil, text=data_args.prompt)
 
     with paddle.no_grad():
         outputs = dino_model(
             image_tensor,
             mask,
-            input_ids=tokenized_out['input_ids'],
-            attention_mask=tokenized_out['attention_mask'],
+            input_ids=tokenized_out["input_ids"],
+            attention_mask=tokenized_out["attention_mask"],
             text_self_attention_masks=tokenized_out[
-                'text_self_attention_masks'],
-            position_ids=tokenized_out['position_ids'])
+                "text_self_attention_masks"],
+            position_ids=tokenized_out["position_ids"], )
 
     logits = F.sigmoid(outputs["pred_logits"])[0]  # (nq, 256)
     boxes = outputs["pred_boxes"][0]  # (nq, 4)
@@ -171,12 +185,12 @@ def main():
         for box, label in zip(boxes, pred_phrases):
             show_box(box, plt.gca(), label)
 
-        plt.axis('off')
+        plt.axis("off")
         plt.savefig(
-            os.path.join(model_args.output_dir, 'mask_pred.jpg'),
+            os.path.join(model_args.output_dir, "mask_pred.jpg"),
             bbox_inches="tight",
             dpi=300,
-            pad_inches=0.0)
+            pad_inches=0.0, )
 
     logger.info("finish!")